Bugged randomizer when upgrading items

curlygirl2000 · May 2016

Yesterday I attempted to upgrade an artifact that was a 10% chance, using preservation wards. After 25 tries, I gave up. I'm a statistics major, and can tell you that the odds of not making a 10% chance after 25 tries are 7.1%. Last week, it took me 15 tries to upgrade an item that was a 50% chance. *15* tries. The odds of that (assuming your randomizer is not broken) is .003%! Many people on our guild website have noticed the same phenomenon. I've also decided to start keeping a record of upgrades chances vs observed actual values, and see exactly what current chances are panning out to be. Of the many upgrades I've kept track of so far, very few happen before the statistical expectation. Most happen after, which of course would be good for the company, but would also show that something is not accurate in the randomizer. Anyone else seeing this?

zibadawa · May 2016

HAMSTER happens.

Everyone thinks whatever game they're playing is lying to them and is out to get them.

None of them are right.

phoenix1021 · May 2016

zibadawa said:
HAMSTER happens.

Everyone thinks whatever game they're playing is lying to them and is out to get them.

None of them are right.

Indeed, the random generators always seem to be out to get you. It's called negativity bias.

Unknown · May 2016

This content has been removed.

ironzerg79 · May 2016

Whether or not it's actually bugged, I personally feel like the "X chance to upgrade" part of this game is one of the most disappointing "features". It honestly feels awful to take all the time and effort to collect everything you need to refine something up to the next level, then repeatedly fail to actually upgrade that item over and over and over...even if you have enough wards.

It's like if a member of your preferred sex asks for a kiss, then slaps you as you lean in. "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap!

Then finally after a getting slapped in the face an undetermined number of times, you finally get the kiss.

It should feel good to progress, but at the end of it you're just left feeling angry and exasperated.

Unknown · May 2016

This content has been removed.

zibadawa · May 2016

That's how RNG works. Someone always gets reamed. If there weren't non-zero chances at arbitrarily long tails, it wouldn't be a binomial random variable. And it of course always frustrates certain players, because most of them aren't properly educated in probability and statistics, and even those who are might be poor losers or otherwise too emotionally stunted to handle random misfortune with any grace or civility. Nevertheless, it is a point of contention in game design of how much you allow your players to be decimated by randomness. I've known games that instituted hard caps on the tails. So if a quest item drops 20% of time, maybe they throw in code which also says it's a 100% drop on the 50th kill if you haven't gotten it yet (or since the last one you got). In this case that's still 10 times the average, but it means every player has a pre-established maximum amount of kills they have to do to get the task completed, with most people being done much faster. And that maximum saves the excruciating frustration for the few players that otherwise would have gone hundreds of kills with nothing to show for it but an aneurysm. Because, statistically speaking, that's going to happen (more than once) to someone eventually during the course of the game's life.

But with no hard caps, you can always find yourself in that long tail, and that's simply the nature of the beast. And when lots of things are random, you'll find yourself in a long tail many many times over the course of your play time. You just might not notice most of them--hence the aforementioned "negativity bias". You're not likely to complain much if for 1000 attacks you've critted 50% of the time when your stat sheet only has you at 35%, and you probably won't even notice if, alternatively, you roll the top 10% on your damage rolls for 1000 attacks in a row. But if you've played regularly for a year or more, you've probably done this at least once. But you're a lot more likely to notice if you're in the bottom 10% for 1000 attacks, or critting only 20% of the time instead of 35%, because you will just suddenly feel like you're inexplicably weak and that's not what you're playing the game for.

dsn1118 · May 2016

zibadawa said:
That's how RNG works. Someone always gets reamed. If there weren't non-zero chances at arbitrarily long tails, it wouldn't be a binomial random variable. And it of course always frustrates certain players, because most of them aren't properly educated in probability and statistics, and even those who are might be poor losers or otherwise too emotionally stunted to handle random misfortune with any grace or civility. Nevertheless, it is a point of contention in game design of how much you allow your players to be decimated by randomness. I've known games that instituted hard caps on the tails. So if a quest item drops 20% of time, maybe they throw in code which also says it's a 100% drop on the 50th kill if you haven't gotten it yet (or since the last one you got). In this case that's still 10 times the average, but it means every player has a pre-established maximum amount of kills they have to do to get the task completed, with most people being done much faster. And that maximum saves the excruciating frustration for the few players that otherwise would have gone hundreds of kills with nothing to show for it but an aneurysm. Because, statistically speaking, that's going to happen (more than once) to someone eventually during the course of the game's life.

But with no hard caps, you can always find yourself in that long tail, and that's simply the nature of the beast. And when lots of things are random, you'll find yourself in a long tail many many times over the course of your play time. You just might not notice most of them--hence the aforementioned "negativity bias". You're not likely to complain much if for 1000 attacks you've critted 50% of the time when your stat sheet only has you at 35%, and you probably won't even notice if, alternatively, you roll the top 10% on your damage rolls for 1000 attacks in a row. But if you've played regularly for a year or more, you've probably done this at least once. But you're a lot more likely to notice if you're in the bottom 10% for 1000 attacks, or critting only 20% of the time instead of 35%, because you will just suddenly feel like you're inexplicably weak and that's not what you're playing the game for.

Do you know what is the statistical chance of getting 20-30 failures one after another with 10% success chance?

dsn1118 · May 2016

To my knowledge you have 50% chance to get 6 failures
35% chance to get 10 failures
20% chance to get 15 failures with 10% success rate.

So I think I have to see 10-15 failures most of the time if the RNG really works.or maybe I am wrong I only took one class about statistics and used it as sleeping chance between classes

lirithiel · May 2016

During this last 2xRP event I attempted to upgrade my lvl 139 Sigil of the Hunter to Mythic. After burning through 62 (yes SIXTY-TWO) Pres Wards I gave up and now my artifact can sit at lvl 139 forever for all I care. RNG is this game is especially borked.

Unknown · May 2016

This content has been removed.

durynte · May 2016

PNRG activity is large part of f2p revenue, so I'd expect there has been quite a bit of thougth put in. Too good the casino rolls on average are bad for business, as are too bad digital dicing results. If you don't expect anything from it, you could find yourself amused on being lucky. If not lucky, well. I don't care...

Some more on what is at work on digital dicing:
http://forum.arcgames.com/neverwinter/discussion/comment/4349782/#Comment_4349782

ironzerg79 said:
It's like if a member of your preferred sex asks for a kiss, then slaps you as you lean in. "Oh, just kidding, you can kiss me"...slap!

That's are rather funny way to describe it! *giggles*

wildfirede · May 2016

reminds me of an old joke
"But sir you must get slapped alot"
"Yes but i got lots of kisses too"

on topic burned my 25 p.ward reserve on a r8 enchant (20% if i remember correctly ) yesterday.

it would be really cool if every failure rose the success chance by 1% or something in that manner.

plasticbat · May 2016

As far as I can tell, going to an isolated instance helps a lot. Opening the AE and Artifact RP boxes in the weekend, I got dramatically a lot more good stuff comparing with a "normal" instance. It was not just happening once in a blue moon. It was about 50 times and with different toons. Before that, I never saw I could get 6 green in a row. I got 6 green in a row about 5 times. Colour vs white was about 40:60. Hack! I even got multiple blues which only happened once in (may be) 150 before. I could be very lucky. However, the luck spread through multiple days, different toon and different time.

wildfirede · May 2016

my bet - server generates a seed for loots in a zone at maintainance and it can be either exploited or severy manipulated by players who use opentillwin tactic. hence you gettinng better luck in zones where the seed is not used by others.

phoenix1021 · May 2016

ironzerg79 said:
Whether or not it's actually bugged, I personally feel like the "X chance to upgrade" part of this game is one of the most disappointing "features". It honestly feels awful to take all the time and effort to collect everything you need to refine something up to the next level, then repeatedly fail to actually upgrade that item over and over and over...even if you have enough wards.

It's like if a member of your preferred sex asks for a kiss, then slaps you as you lean in. "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap!

Then finally after a getting slapped in the face an undetermined number of times, you finally get the kiss.

It should feel good to progress, but at the end of it you're just left feeling angry and exasperated.

The logical solution would be to make it a 100% chance, but increase the materials required. Who would like to use 100 greater marks of potency and 100 same level enchants to upgrade that enchant to greater?
I just don't see a better way...

beckylunatic · May 2016

dsn1118 said:

zibadawa said:
That's how RNG works. Someone always gets reamed. If there weren't non-zero chances at arbitrarily long tails, it wouldn't be a binomial random variable. And it of course always frustrates certain players, because most of them aren't properly educated in probability and statistics, and even those who are might be poor losers or otherwise too emotionally stunted to handle random misfortune with any grace or civility. Nevertheless, it is a point of contention in game design of how much you allow your players to be decimated by randomness. I've known games that instituted hard caps on the tails. So if a quest item drops 20% of time, maybe they throw in code which also says it's a 100% drop on the 50th kill if you haven't gotten it yet (or since the last one you got). In this case that's still 10 times the average, but it means every player has a pre-established maximum amount of kills they have to do to get the task completed, with most people being done much faster. And that maximum saves the excruciating frustration for the few players that otherwise would have gone hundreds of kills with nothing to show for it but an aneurysm. Because, statistically speaking, that's going to happen (more than once) to someone eventually during the course of the game's life.

But with no hard caps, you can always find yourself in that long tail, and that's simply the nature of the beast. And when lots of things are random, you'll find yourself in a long tail many many times over the course of your play time. You just might not notice most of them--hence the aforementioned "negativity bias". You're not likely to complain much if for 1000 attacks you've critted 50% of the time when your stat sheet only has you at 35%, and you probably won't even notice if, alternatively, you roll the top 10% on your damage rolls for 1000 attacks in a row. But if you've played regularly for a year or more, you've probably done this at least once. But you're a lot more likely to notice if you're in the bottom 10% for 1000 attacks, or critting only 20% of the time instead of 35%, because you will just suddenly feel like you're inexplicably weak and that's not what you're playing the game for.
Do you know what is the statistical chance of getting 20-30 failures one after another with 10% success chance?

Whatever it is, it probably balances out with the number of times I've hit a 10% chance success in 3 tries or fewer.

ironzerg79 · May 2016

magenubbie said:

ironzerg79 said:
It's like if a member of your preferred sex asks for a kiss, then slaps you as you lean in. "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap!
Very accurate analogy

I really haven't upgraded much past R9/R10 because this system stresses me out so much. Probably irrationally so. But I hate it. I just hate clicking the combine button and seeing Fail, Fail, Fail, Fail, Fail...I have more than enough AD to actually upgrade, I just hate the system.

Fail. Fail. Fail. Slap. Slap. Slap.

See? I'm getting totally irrational right now, but I can't help it.

Slap. Slap. Slap.

Just thinking about pushing that button to upgrade.

Slap!

beckylunatic · May 2016

ironzerg79 said:

magenubbie said:

ironzerg79 said:
It's like if a member of your preferred sex asks for a kiss, then slaps you as you lean in. "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap! "Oh, just kidding, you can kiss me"...slap!
Very accurate analogy

I really haven't upgraded much past R9/R10 because this system stresses me out so much. Probably irrationally so. But I hate it. I just hate clicking the combine button and seeing Fail, Fail, Fail, Fail, Fail...I have more than enough AD to actually upgrade, I just hate the system.

Fail. Fail. Fail. Slap. Slap. Slap.

See? I'm getting totally irrational right now, but I can't help it.

Slap. Slap. Slap.

Just thinking about pushing that button to upgrade.

Slap!

"*#%$ you Cryptic, you're ruining my life."

While I make fun of the out-to-get-me attitude, there is a lot in the game that really does inspire that kind of feeling. Just because something actually works right with a large enough sample size, doesn't mean it doesn't feel horrible when you're on the bad end of the bell curve.

Makes the addition of the Paranoid Delusion oddly meta. And it looks. Just. Like. Us.

sorce · May 2016

lirithiel said:
During this last 2xRP event I attempted to upgrade my lvl 139 Sigil of the Hunter to Mythic. After burning through 62 (yes SIXTY-TWO) Pres Wards I gave up and now my artifact can sit at lvl 139 forever for all I care. RNG is this game is especially borked.

Last double RP I tried to upgrade my Sigil of the Hunter to Mythic - in the end it took 128 pres wards. Starting to think the HR sigil is cursed. :frowning:

ironzerg79 · May 2016

I agree. Tt doesn't have anything to do with the odds or feeling like the system is broken. When you get enchants up to a certain rank, and it's 5% or 3% to succeed, even if every single time it only took me an average of 20 or 33 tries to get it right, watching 19 fails in a row isn't engaging.

It just doesn't feel good, that's all I'm saying.

Getting to the point where you can upgrade your stuff should be the hard part of the journey. Finally getting to hit that "upgrade" button, watching the screen flash and your shiny new item pop up should be your reward. And it should feel awesome.

But the way the system is set up with all the fails in between you hitting that button, and your shiny upgrade popping up sucks. Even if the random chance is perfectly distributed.

It should feel awesome, but it doesn't. I don't imagine many people like failing simply do to pure chance, but we'll all accept a little bit of that. But when the refinement system slaps you in the face with it over and over and over and over, stacking fail after fail in quick succession until it finally succeeds (again, even under the assumption the random chance is 100% totally fair), it just doesn't feel like winning.

Getting gear upgrades in MMOs should feel like winning.

oliboyph · May 2016

Most programs do not have genuine random generators. True random generators use atmospheric noise measures or even radio active isotopes. The truth is, if you completely depend on software, you can't achieve true randomness. That's why it's always possible to recreate a string of randomly generated numbers. Which should not happen if you it is really random.

If you feel like the string of numbers you are generating is repeating against your favor, you should stop and try again later. It's the best advice I can give.

curlygirl2000 · May 2016

Thank you all for your comments, at least, while I do feel on the very loooooong end of the tail in this game, at least I know I'm in good company.

Curlygirl

zibadawa · May 2016

oliboyph said:
Most programs do not have genuine random generators. True random generators use atmospheric noise measures or even radio active isotopes. The truth is, if you completely depend on software, you can't achieve true randomness. That's why it's always possible to recreate a string of randomly generated numbers. Which should not happen if you it is really random.

If you feel like the string of numbers you are generating is repeating against your favor, you should stop and try again later. It's the best advice I can give.

Those are immensely impractical for gaming purposes, and if we want to go to that sort of level I could point out that the current laws of physics are all deterministic, so the randomness is basically just an illusion (quantum wave forms remain probabilistic things, but their evolution is deterministic; and depending on the interpretation for QM you pick you can conclude that the wave forms are just an illusion, as well).

The periodicity of a Mersenne twister is extremely large for practical purposes, and while it is in principle possible to gain complete predictive capacity over the algorithm from a comparatively small sample of consecutive outputs, it's not really feasible or reliable to get that many consecutive results from a single information gatherer in an online game--there will always be some random HAMSTER "interrupting" your sequence with their own activities.

greywynd · May 2016

First attempt: 10% chance
Second attempt: 10% chance
Third attempt: 10% chance
...and so on

Each and every attempt is independent of the preceding attempts. They exist in a vacuum.

Unknown · May 2016

This content has been removed.

thefabricant · May 2016

desisti said:
RNG is extremely streaky... EXTREMELY... I managed to upgrade my DC sigil to mythic after "mere" 44 attempts (majority of friends I know succeeded after 40-50 tries, unlucky ones 50-60). But not only there you can see RNG at work. Professions also fall under the same spell. On alchemy masterwork i managed to fail 32 times in a row trying to achieve R3 reward with 20% success rate... 32 times... Can you calculate odds of that happening

0.8^32 = 0.079228162% chance of happening. To put it another way, out of every 10000 people, you are one of the 8 people this happens to.

Unknown · May 2016

This content has been removed.

dfnce · May 2016

Forget about percentage on screen, use small populated area in low activity period, Monday morning and you will do fine.

It doesn't really matter 50% or 10%, because you should be ready to spend 10-20 wards anyway. But if you fail more than 20 in row, it means you should stop, period.

Why is that, did some one saw the code doesn't really matter. If their server code happen to run on virtualized headless environment without extra entropy sources, their random is barely "true random". Just like majority.

oliboyph · May 2016

zibadawa said:

oliboyph said:
Most programs do not have genuine random generators. True random generators use atmospheric noise measures or even radio active isotopes. The truth is, if you completely depend on software, you can't achieve true randomness. That's why it's always possible to recreate a string of randomly generated numbers. Which should not happen if you it is really random.

If you feel like the string of numbers you are generating is repeating against your favor, you should stop and try again later. It's the best advice I can give.
Those are immensely impractical for gaming purposes, and if we want to go to that sort of level I could point out that the current laws of physics are all deterministic, so the randomness is basically just an illusion (quantum wave forms remain probabilistic things, but their evolution is deterministic; and depending on the interpretation for QM you pick you can conclude that the wave forms are just an illusion, as well).

The periodicity of a Mersenne twister is extremely large for practical purposes, and while it is in principle possible to gain complete predictive capacity over the algorithm from a comparatively small sample of consecutive outputs, it's not really feasible or reliable to get that many consecutive results from a single information gatherer in an online game--there will always be some random HAMSTER "interrupting" your sequence with their own activities.

Yes, but because of the way computers generate numbers it's the only way to make real random numbers. Example this is the cheapest source of real random generators on the net. https://www.random.org/

Unknown · May 2016

This content has been removed.

Bugged randomizer when upgrading items

Comments